Abstract
Abstract 2603
Although CBF-AML (i.e. with t(8;21) or inv(16)/t(16;16)) represents a favorable cytogenetic AML subgroup (Döhner, Blood 2010), 35–45% of these patients still relapse after standard intensive chemotherapy. The immunoconjugate gemtuzumab ozogamicin (GO) was shown to be effective in patients with relapsed AML in non randomized studies and has been recently demonstrated in a Phase 3 trial as associated with a significant benefit in younger adults with CBF-AML (Burnett, JCO 2011). In this study, we thus investigated the impact of GO-based salvage at first relapse in this specific subgroup of patients with CBF-AML.
We retrospectively analysed the medical records of 84 patients aged 60 years or less with CBF-AML in first relapse after intensive chemotherapy and treated in 18 French centers. None of these patients received allogeneic (alloSCT) or autologous (autoSCT) hematopoietic stem cell transplantation in first complete remission (CR). As salvage, 27 patients received GO, combined with high-dose cytarabine in most of them; 21 patients received high-dose cytarabine and anthracycline without GO; 36 patients received conventional chemotherapy based on standard-dose cytarabine and anthracycline. Post-remission therapy was alloSCT in 49 patients, autoSCT in 17 patients, and chemotherapy alone in 11 patients.
Among 84 patients with a median age of 39 years [16–58], 36 patients had t(8;21) AML and 48 patients had inv(16)/t(16;16) AML. Median CR1 duration was 12.9 months [2.6–55.3]. Second complete remission (CR2) rate was 92% (77/84), and early death rate was 1% (1/84). The median follow up was 4.0 years. The 5-year overall survival (5y-OS) and relapse-free survival (5y-RFS) was 52% [39–64%] and 48% [36–60%] respectively. Patients receiving alloSCT in CR2 had a better outcome (5y-OS, 56% versus 43%; p=0.05). In patients not allografted in CR2, RFS was similar after autoSCT and chemotherapy alone (5y-RFS, 44% versus 47%, respectively). Patients treated with GO had similar CR rate but a lower risk of second relapse and a better survival than other patients (5y-RFS, 89% versus 55%; p=0.05 and 5y-OS, 90% versus 45%; p=0.03). In univariate analysis, other factors associated with a better OS were younger age, longer CR1 duration, but not CBF subtype (p=0.03, 0.01, and 0.20, respectively). In multivariate analysis adjusted on age, CR1 duration, and CBF subtype, GO salvage was still associated with a significant benefit in OS (HR=0.16 [0.04–0.69], p=0.01) and RFS (HR=0.19 [0.04–0.80], p=0.02). With a median post-relapse follow-up of 2.2 years, no relapse nor death were observed in the 19 patients who received GO salvage followed by alloSCT in CR2 (p=0.007 for RFS; p=0.008 for OS). Moreover, in patients who received alloSCT, previous GO therapy significantly improved post-transplantation outcome.
Younger patients with CBF-AML in first relapse had a high second complete remission rate regardless the intensive chemotherapy salvage. More interestingly, the outcome of these patients was significantly improved by the addition of GO-based salvage, especially when followed by alloSCT.
Off Label Use: GO is available in Europe as a compassionate treatment for relapsed AML.
Author notes
Asterisk with author names denotes non-ASH members.